@ubnt/cawcut@0.3.0
CawCut developer CLI — auth, generate, app run, task status
AI Security Review
scanned 10d ago · by lpm-firewall-aiThe package automatically installs first-party agent skills during global installation and may bootstrap them on CLI use, creating an unresolved consent risk. The inspected source does not establish a concrete malicious attack, so this warrants a warning rather than a block.
Decision evidence
public snapshot- package.json runs dist/postinstall.js after installation.
- The postinstall invokes skill installation for global installs.
- The installer runs npx skills add globally with automatic confirmation for CawCut skills and defaults to Cursor, Claude Code, and Codex.
- The default skill source is cawcut/skill, and the installer targets agent skill directories under the user's home directory.
- The CLI can also install the skills on a regular first invocation when they are missing.
- The postinstall exits unless this is a global install and skips setup under CI or when CAWCUT_SKIP_SETUP is set.
- The inspected code shows setup of named skills in skill directories; it does not show credential collection, exfiltration, or broad agent configuration changes.
Source & flagged code
8 flagged · loading sourcePackage defines install-time lifecycle scripts.
package.jsonView on unpkgInstall-time lifecycle script is not statically allowlisted and needs review.
package.jsonView on unpkgPackage source references child process execution.
dist/lib/uninstall-cawcut.jsView on unpkg · L1Findings
Affected versions and remediation
This report applies to @ubnt/cawcut@0.3.0.
See all versions of @ubnt/cawcut for other recorded verdicts.
- Review the evidence and your use of @ubnt/cawcut@0.3.0 before allowing it. Restrict the permissions described in this report.
- Choose an independently verified alternative or release. This report does not establish that other versions are safe.
Evidence last updated: .
Think this verdict is wrong? Dispute it on GitHub or read how reports are produced.